Experimental investigation of a recursive modelling MPC system for space heating within an occupied domestic dwelling

Rogers, Dan, Foster, Martin and Bingham, Chris (2014) Experimental investigation of a recursive modelling MPC system for space heating within an occupied domestic dwelling. Building and Environment, 72 . pp. 356-367. ISSN 0360-1323

12839 bingham-rogers_2014.pdf

Request a copy
[img] PDF
12839 bingham-rogers_2014.pdf - Whole Document
Restricted to Repository staff only

Item Type:Article
Item Status:Live Archive


A key contributor to excessive energy usage in UK domestic dwellings is the poor temperature regulation of fluid filled heat emitters that form the most common constituent of central heating systems. This is in-part due to oversizing of the heat emitters, and deficiencies in localised closed-loop temperature control. In an attempt to address this, Controllable Radiator Valves (CRVs) are becoming an increasingly popular domestic technology with the ability to allow previously unrealisable control schemes to operate the traditional central heating system with minimal mechanical modification. Following previous reported work, therefore, this paper presents a new family of Recursive Modelling Model Predictive Controllers (RM-MPCs) for use with low cost thermic CRVs. The ability of the presented control methodologies to maintain superior temperature regulation despite the use of oversized heat emitters, is a key contribution of the paper. Furthermore, unlike previously reported techniques, the underlying recursive modelling method has been reformulated so that traditional parameter matching calculations do not now require a computationally intensive curve fitting stage. A comparison of techniques is included using experimental measurements from both an oversized oil filled heat emitter within a test chamber, and also from BS EN 442 water-filled heat emitters within an occupied dwelling. Results show the proposed methodologies can be realised using more cost-effective thermoelectric valves, whilst providing superior set point tracking. © 2013 Elsevier Ltd.

Keywords:Domestic, HVAC, Model predictive control, Modelling, NotOAChecked
Subjects:K Architecture, Building and Planning > K210 Building Technology
H Engineering > H221 Energy Resources
Divisions:College of Science > School of Engineering
Related URLs:
ID Code:12839
Deposited On:06 Jan 2014 11:14

Repository Staff Only: item control page